Grants:PEG/Wikimedians in California/Wikimedia California

Not funded

This grant submission from an group, organization, or individual in 2013-14 was not funded by the Wikimedia Foundation.

Prior to 2011, reasons for not funding this submission were communicated in private and may not be available on the discussion page. For more information about a particular request, please contact the grantee directly or send your questions to grants at wikimedia dot org.
Beginning in 2011, reasons for not funding this submission are communicated publicly on the discussion page of this submission.
Since this grant request was not funded, no report is required and no further action is required from the grantee.

Note: The Wikimedia California chapter is not yet formed. The chapter grant process, "is open to all chapters officially recognized by the Wikimedia Foundation. If your chapter is still in development, you can still apply for funds (especially when they are relevant to getting your chapter off the ground), and we will request input from the chapters committee to help us assess whether your chapter is ready to absorb them." -- http://meta.wikimedia.org/wiki/Wikimedia_chapters/WMF_grants

Current plans are to also apply for grants such as [1][2] and [3] and return as much Foundation money as possible from awards such as those.

Legal name of chapter: Wikimedia California, Inc.
Grant contact name: James Salsman
Grant contact e-mail: jsalsman@talknicer.com
Grant contact title (position): volunteer
Project lead name(s): to be determined
Full project name: Initial chapter software, hardware, and Wikinews, event, and outreach support
Amount requested (in USD): $550,000
Provisional target start date: June, 2010
Provisional completion date: September, 2010 - June, 2011 (various completion dates; some projects potentially ongoing)

Software projects

There were ten Google Summer of Code 2010 student applicants whose projects were scored positively by mentors, but rejected because the Foundation did not request enough slots.

There are also several volunteer mentors who are all still unassigned to any student projects. Generally, we should hire un-placed volunteer mentors and provide the $5,000/three months stipend recommended by Google.

We also want to establish ongoing projects, for example hiring Mediawiki volunteers and http://odesk.com workers to go through and work on the Bugzilla queue. Also we might do a peer-to-peer wiki edit conflict resolution project using funds through Amazon Mechanical Turk.

Software budget breakdown

Total: $150,000
- Student stipends: $50,000 (10 x $5,000/three months)
- Mentor fees: $50,000 (estimated and subject to negotiation)
- Mediawiki Bugzilla queue work and management via http://odesk.com : $50,000 initially; subject to renewal if successful in the Foundation's judgment

Software project scope

We will try to emulate the Google Summer of Code process for the submitted student projects which were scored positively but not allocated slots. We will try to reimburse mentors for their help, and also ask them (if they have free time) to work the Mediawiki Bugzilla queue through the http://odesk.com software contracting system.

Software project goal

We hope to mentor ten students and provide at least 1,500 hours of Mediawiki bug fixes and patch submission. If time and money permit, we will pilot a distributed edit conflict resolution system to make peer-to-peer wikis a reality.

Non-financial requirements -- software

We will need Tim to approve check-in access for participants if they don't have it already. We would also like Jay's help with a press release, once the project is complete.

Fit to strategy -- software

The proposed projects have been designed to increase reach and participation (e.g., by displaying the most popular related articles), quality (e.g. interactive instructional quiz content importation from widely-used formats), credibility (e.g. by increasing audio recording coverage in the (non-English) Wiktionaries), organizational maturity and effectiveness (by exploring a payments system for bugzilla-based workers), and financial sustainability (we plan to apply for NSF grants to be self-sufficient and return the money to the Foundation.)

Other benefits -- software

Many volunteers believe that the Foundation is not adding software or support capacity in proportion to its financial growth. This will help relieve some of that pressure.

Measures of success -- software

We will consider these projects a success if they provide their specified deliverables. For project deliverables in the following student proposals, please see: Cam Vilay, Damon Wang, Johan Gunnarsson, Laszlo Kozma, Meadowlark Bradshaw (note: Meadowlark was accepted by Berkeley and may have transfer-related classwork over the Summer), Maciej Sawicki, Michael A. White, Neeraj Agarwal, Ryan Miller, and Shubhandra Singh.

Maciej has taken a regular job over the Summer. Meadowlark was accepted in to Berkeley, and mentor Yaron has offered to do his project for less than half of what it would have cost Google.

Hardware, colocation, and internet service

In order to purchase IBM's latest servers with the supposed best dhrystones/$ we will need about $20,000 per each, plus rack space and internet somewhere. However, we should leave this open-ended for the Foundation to decide on the exact amount they want to spend for external hardware support.

Hardware budget breakdown

to be determined (estimate: $150,000)

Database and squid servers
- about $20,000 each for IBM System x3850/3950 M2[4]
  - 7233-6LY with 256GB
  - 7233-4MY with 256GB
  - 7233-6MY (part number: 72336MG) with 256GB
- blade systems with larger processor caches[5]
  - e.g 7872-82Y
Image bundle and backup servers
- the best two of the above servers
- plus FusionIO solid state storage
- plus 2 TB hard drives
- plus enough network-attached storage to serve the commons, enwiki, and other project image bundles (10 TB?[6])
Colocation facilities
Internet service

Hardware project scope

We will attempt to produce a mirror of the highest-volume Foundation servers with the best price/performance ratio from available hardware.

Hardware project goal

The goal of this project is to empirically measure the price/performance ratio of latest-generation server (and, if budget and time permits, peer-to-peer) hardware suitable for the Foundation's mission.

Non-financial requirements -- hardware

We may need help with mirroring or otherwise forking broken dumps, such as image bundles and the other currently broken dumps.

Fit to strategy -- hardware

The project will provide previously unavailable, useful data for Foundation hardware purchasing decisions.

Other benefits -- hardware

This will allow us to create a Foundation mirror which could serve as a fail-over in case of datacenter outages, and as a testbed for a third-party edit conflict resolution systems such as are necessary for peer-to-peer wiki software.

Measures of success -- hardware

We will consider this project a success when we have useful empirical statistics on the price/performance ratio of next-generation hardware after a reasonably complete mirror has been established.

Wikinews

Recommendations:

journalism awards
travel expenses
reporters
fact checkers
photographers
camera and recording equipment
proofreaders
editors (can probably depend on volunteers 100% for editors not including professional proofreaders)
initial request $250,000 or Foundation's option; Jon is the contact for this
note: Wikinews is required to be independent, which should mean that the success criteria would need to be determined third-party journalism experts.

Wikinews budget breakdown

to by determined (by Jon or other accredited Wikinews staff)

Wikinews project scope

Acting in accordance with the Wikinews mission, to collect and disseminate news.

Wikinews project goal

to be determined: As an independent investigative journalism institution, the goals should be set by Wikinews staff

Non-financial requirements -- Wikinews

unknown at present

Fit to strategy -- Wikinews

This will help Wikinews expand its reach and utility. Experts contacted have been skeptical of the viability and utility of all-volunteer news gathering and reporting, as have accredited Wikinews volunteers.

Measures of success -- Wikinews

We will consider this project a success if it is judged so by an independent panel of journalism experts.

Maker Faire

We need to get back-up support for things which had thought to be likely borrowable but might need to be bought by someone on short notice. See also w:Wikipedia:Meetup/San_Francisco/Maker_Faire_2010

Request: No more than $5,000 contingent on actually needing any of it. We'll see what happens.

Scope: Hardware as specified by volunteers at the coordination meetings, e.g., five laptops and/or component systems, some of which are intended for public access, [7]x1, and [8]x2

Rationale: Outreach

Success criteria: Volunteers not having to make last-minute hardware purchases for which they aren't reimbursed.

West Coast Wikiconference

Initially this is being attempted without any hosting, travel or housing stipends or scholarships, but that could change. See West Coast WikiConference.

Initial request: to be determined

Provide a means to disseminate knowledge about the Chapter and Wikimedia in general.
Mailing List Delivery?
- Open to Everyone Done

Does this need to be budgeted?

Releasing Knowledge

Establish working relationships with museums and archives
- Initial contact
- Follow ups
- Appreciation Letters
- Press Releases
Goals
- Gain access for photographers to museum collections
- Gain access to private archives for information